Using Sentence-Level LSTM Language Models for Script Inference

TL;DR

This paper compares traditional structured event models with recent RNN-based models that predict sentences directly from raw tokens, finding comparable performance in script inference tasks.

Contribution

It demonstrates that RNN language models operating on raw tokens perform similarly to structured models on event prediction, highlighting their potential for script inference.

Findings

RNN models are roughly comparable to structured event models in predicting missing events.

Direct token-based models can match the performance of structured verb-argument systems.

The study provides insights into the effectiveness of neural language models for script inference.

Abstract

There is a small but growing body of research on statistical scripts, models of event sequences that allow probabilistic inference of implicit events from documents. These systems operate on structured verb-argument events produced by an NLP pipeline. We compare these systems with recent Recurrent Neural Net models that directly operate on raw tokens to predict sentences, finding the latter to be roughly comparable to the former in terms of predicting missing events in documents.